Improving (Q)SAR predictions by examining bias in the selection of compounds for experimental testing

ABSTRACT

Existing data on structures and biological activities are limited and distributed unevenly across distinct molecular targets and chemical compounds. The question arises if these data represent an unbiased sample of the general population of chemical-biological interactions. To answer this question, we analyzed ChEMBL data for 87,583 molecules tested against 919 protein targets using supervised and unsupervised approaches. Hierarchical clustering of the Murcko frameworks generated using Chemistry Development Toolkit showed that the available data form a big diffuse cloud without apparent structure. In contrast hereto, PASS-based classifiers allowed prediction whether the compound had been tested against the particular molecular target, despite whether it was active or not. Thus, one may conclude that the selection of chemical compounds for testing against specific targets is biased, probably due to the influence of prior knowledge. We assessed the possibility to improve (Q)SAR predictions using this fact: PASS prediction of the interaction with the particular target for compounds predicted as tested against the target has significantly higher accuracy than for those predicted as untested (average ROC AUC are about 0.87 and 0.75, respectively). Thus, considering the existing bias in the data of the training set may increase the performance of virtual screening.     

The effect of the sintering atmosphere on the densification of B4C ceramics

Densification of boron carbide during sintering may be improved by a two-stage process, namely heating to 2000°C under vacuum and sintering at 2190°C under argon. This sintering regime allows achieving a relative density of the ceramic bodies fabricated from a fine powder higher than 95%. The nitrogen treatment of the boron carbide phase at 1900°C leads to the formation of the BN phase and precipitation of graphite. Vacuum treatment of these samples at 2000°C leads to decomposition of the boron nitride phase. The liberated free boron may again react with graphite to form in situ boron carbide particles. The experimental investigations of the sintering behavior of the boron carbide phase under various atmospheres supported the thermodynamic predictions regarding the phase transformation. No evidence, however, was found for enhanced sintering under a nitrogen atmosphere.     

Nonlocality as a feature of the physical world

Aprediction puzzle leads to a form ofnecessary realism which forces the rejection of a central tenet of the Copenhagen interpretation. This leads to reconsidering conceptual difficulties related to Bell's locality premise. It is shown that aparadox of elementary probability theory puts new light on Bell's assumption that causality and statistical independence are mutually incompatible. Nemo dat quod non habet : No one can offer more than he has ability to give; in that spirit, this paper is dedicated to the memory of J.S. Bell.     

Application of torsion angle molecular dynamics for efficient sampling of protein conformations

We investigate the application of torsion angle molecular dynamics (TAMD) to augment conformational sampling of peptides and proteins. Interesting conformational changes in proteins mainly involve torsional degrees of freedom. Carrying out molecular dynamics in torsion space does not only explicitly sample the most relevant degrees of freedom, but also allows larger integration time steps with elimination of the bond and angle degrees of freedom. However, the covalent geometry needs to be fixed during internal coordinate dynamics, which can introduce severe distortions to the underlying potential surface in the extensively parameterized modern Cartesian-based protein force fields. A "projection" approach (Katritch et al. J Comput Chem 2003, 24, 254-265) is extended to construct an accurate internal coordinate force field (ICFF) from a source Cartesian force field. Torsion crossterm corrections constructed from local molecular fragments, together with softened van der Waals and electrostatic interactions, are used to recover the potential surface and incorporate implicit bond and angle flexibility. MD simulations of dipeptide models demonstrate that full flexibility in both the backbone phi/psi and side chain chi1 angles are virtually restored. The efficacy of TAMD in enhancing conformational sampling is then further examined by folding simulations of small peptides and refinement experiments of protein NMR structures. The results show that an increase of several fold in conformational sampling efficiency can be reliably achieved. The current study also reveals some complicated intrinsic properties of internal coordinate dynamics, beyond energy conservation, that can limit the maximum size of the integration time step and thus the achievable gain in sampling efficiency.     

Retention in coupled capillary column supercritical fluid chromatography with lipids as model compounds

Summary We have investigated to which extent retention data, acquired on single capillary columns, can be used for predicting retention factors in a coupled column system. For this purpose we utilized a model mixture of 18 lipid components with widely different vapor pressures and polarities. The sample was chromatographed on two columns, SB-biphenyl-30 (70% methyl-30% biphenylpolysiloxane) and SB-cyanopropyl-50 (50% methyl-50% cyanopropylsiloxane). Experimental retention factors, acquired in coupled column systems with two columns connected in different order, were thus compared with values calculated from runs on each single column. The agreement between calculated and experimental values generally was better than 5% without any pressure drop correction.To study the possibility of predicting retention behavior in a wide pressure range from a limited number of experiments, we also investigated the relation between solute retention and mobile phase density. We found that all data could be fitted to second order equations, which gives the possibility to optimize the resolution with respect to pressure from a limited number of runs at different pressures.     

Ab initio prediction of vibrational spectra: A database approach

Efforts to develop a database of quadratic force fields for organic molecules are described. The database is based on systematic ab initio calculations, scaled to reproduce the experimentally observed frequencies. The choice of the theoretical method, the basis sets, geometries, internal coordinates and the scaling procedure are discussed. A key point in the procedure is the automatic generation of the internal valence coordinates. This is also very advantageous for geometry optimization. The database should permit the prediction of vibrational frequencies for most organic molecules to 10–20 cm⁻¹, together with semiquantitative intensities. The accuracy is sufficient to identify unknown compounds from a list of reasonable candidates.     

Statistical inference and modelling of momentum in stock prices

The following results are obtained, (i) It is possible to obtain a time series of market data {y(t)} in which the fluctuations in fundamental value have been compensated for. An objective test of the efficient market hypothesis (EMH), which would predict random correlations about a constant value, is thereby possible, (ii) A time series procedure can be used to determine the extent to which the differences in the data and the moving averages are significant. This provides a model of the form y(t)-y(t-l)=0.5{y(t- l)-y(t-2)}+ε(t)+0.8ε(r-1) where ε(t) is the error at time t, and the coefficients 0.5 and 0.8 are determined from the data. One concludes that today's price is not a random perturbation from yesterday's; rather, yesterday's rate of change is a significant predictor of today's rate of change. This confirms the concept of momentum that is crucial to market participants. (iii) The model provides out-of-sample predictions that can be tested statistically. (iv) The model and coefficients obtained in this way can be used to make predictions on laboratory experiments to establish an objective and quantitative link between the experiments and the market data. These methods circumvent the central difficulty in testing market data, namely, that changes in fundamentals obscure intrinsic trends and autocorrelations. This procedure is implemented by considering the ratio of two similar funds (Germany and Future Germany) with the same manager and performing a set of statistical tests that have excluded fluctuations in fundamental factors. For the entire data of the first 1149 days beginning with the introduction of the latter fund, a standard runs test indicates that the data is 29 standard deviations away from that which would be expected under a hypothesis of random fluctuations about the fundamental value. This and other tests provide strong evidence against the efficient market hypothesis and in favour of autocorrelations in the data. An ARIMA time series finds strong evidence (9.6 and 21.6 standard deviations in the two coefficients) that the data is described by a model that involves the first difference, indicating that momentum is the significant factor. The first quarter's data is used to make out-of-sample predictions for the second quarter with results that are significant to 3 standard deviations. Finally, the ARIMA model and coefficients are used to make predictions on laboratory experiments of Porter and Smith in which the intrinsic value is clear. The model's forecasts are decidedly more accurate than that of the null hypothesis of random fluctuations about the fundamental value.     

考虑防御系统延迟使其预测误差最大的突防弹道设计

对中段多脉冲机动突防弹道的设计问题进行了研究。采用的原理与方法有别于成熟的航天器多脉冲变轨,而是基于路径规划的思想对弹道设计问题简化。考虑敌方防御系统延迟,改进多脉冲点火模型,并基于变射面的思想对关机点参数进行了设计及优化。然后,从使敌方预测误差最大的角度提出评价函数,并利用遗传算法进行优化。最后,给出了一种同时满足导弹机动突防与打击精度要求的多脉冲弹道设计方法。从模型的可行性,方法的有效性、灵活性、迭代效率及精度等方面进行了仿真验证分析。在PC机仿真中,20 s内就设计出一条保证打击精度的中段突防弹道。结果表明,建立的模型是可行的,提出的突防弹道设计方法是快速有效的。